Over the past decade, the electoral advances of far-right and nationalist actors have made “new nationalism” a central object of empirical and theoretical inquiry. Much of the resulting scholarship has focused on exclusionary and ethnocentric mobilisation. While these studies offer interesting insights on far-right nationalism in particular, nationalism is often argued to have a Janus-faced quality; nationalism can endorse restrictive boundaries of nationhood and hostility towards outsiders, but it can also underpin civic-democratic projects grounded in citizenship, values, and self-determination. This duality foregrounds a core comparative puzzle: why do some societies gravitate toward exclusionary variants of nationalism while others sustain more inclusive ones, and how do similar nationalist agendas and frames appear to spread across countries within relatively short periods of time?
Existing research leaves three key gaps. First, much work focuses predominantly on exclusionary expressions of nationalism, capturing only part of the phenomenon and offering limited leverage for explaining why inclusive variants persist or become dominant in some contexts. Second, scholarship remains dominated by single-country studies or regionally bounded comparisons (e.g., within Europe), limiting our ability to identify generalisable patterns and to distinguish transnational diffusion from parallel domestic developments. Third, many accounts treat nationalism primarily as an elite-led project, leaving under-specified the communicative mechanisms through which nationalist meanings gain resonance among mass publics. This project addresses these gaps by advancing a communication-centred, transnational approach that conceptualises nationalism as mediated through digital news: nationalism-relevant meanings circulate through cross-border flows, are domesticated into local interpretive frameworks, and influence public attitudes via media effects. These questions are especially salient for Taiwan, where competing understandings of nationhood coexist and where public debate unfolds within overlapping cross-Strait and global information environments.
The project consists of three work packages (WPs). WP1 maps the global diffusion and domestication of media content relevant to nationalism using a comparative dataset of 100,000+ news articles from 270 outlets across 90 World Values Survey countries and deploys cutting-edge computational methods, a retrieval-augmented generation-based (RAG) large language model classifier, to identify frames and issues at scale. WP2 examines media effects in Taiwan through a two-wave panel survey (N=300) combined with privacy-preserving digital data donation (browsing and YouTube histories), linking individual exposure to WP1 content measures, and estimating selective exposure and media effects via structural equation modelling. WP3 synthesises findings to develop a transnational model of nationalism that specifies how nationalist meanings circulate, are recontextualised, and gain public resonance across diverse societies.